Output power forecasting for 2kW monocrystalline PV system using response surface methodology

Upkli, Wenny Rumy and Wan Abdul Razak, Intan Azmira and Azmi, Aimie Nazmin and Ab Rahman, Azhan and Bohari, Zul Hasrizal (2019) Output power forecasting for 2kW monocrystalline PV system using response surface methodology. International Journal of Electrical Engineering and Applied Sciences (IJEEAS), 2 (2). pp. 23-32. ISSN 2600-7495

Text 5076-15356-1-PB.PDF Download (1MB)

Abstract

Photovoltaic (PV) system is a renewable energy source that not only able to reduce the effect of greenhouse gas towards the environment, but also a highly profitable industry nowadays. To determine the Return of Investment (ROI) of a newly installed system, forecasting is crucial. Thus, the purpose of this study is to produce a prediction model for the yearly output power of the PV system using three environmental elements; irradiance, back module temperature and ambient temperature by Response Surface Methodology (RSM). To do so, MATLAB RStool which is consisting of four models; multiple linear regression (MLR), interaction, pure quadratic, and full quadratic were used. The 5 minute sampling size of year 2014 weather station data of the three environmental elements and output power of a 2kW Monocrystalline real PV system were used for training. Whereas, year 2015 data of the aforementioned elements were used for validation. The coefficient of determination (R2) method and root mean square error (RMSE) approach were used to determine the most accurate prediction model. Results shown that, full quadratic is the most accurate prediction model with R2 value of 0.9995 and RMSE of 8%. It is hoped that the prediction model introduced can be a viable method to be used by the PV system installer.

Item Type:	Article
Uncontrolled Keywords:	Ambient temperature, Back panel temperature, Irradiance, Photovoltaic (PV), RSM
Divisions:	Faculty of Electrical Engineering
Depositing User:	Sabariah Ismail
Date Deposited:	14 Sep 2021 17:14
Last Modified:	21 Jul 2023 15:34
URI:	http://eprints.utem.edu.my/id/eprint/25301
Statistic Details:	View Download Statistic

Actions (login required)

View Item